diy solar

diy solar

Sizing calculations

MapleGuy

New Member
Joined
Jan 17, 2023
Messages
16
Location
Stafford Springs, CT
Let me start by saying that I don't have any experience with home solar systems but I do have experience with using solar systems off-grid in an RV. Before I move onto designing a system I want to make sure my calculations are correct. We were originally going to be building a home but our plans have changed and we are looking at buying an existing home and I'd still like to supply most of my own power.

If the new house used 600KWh a month, that's 20kw/day or drawing 90-160 amps from the batteries, correct? I'm asking because if I look at charge time calculator it looks like I would be able to recharge that in 1.66 peak hours with a 1200 watt solar panel array. That is based on what I know from the RV world but when I look at system sizing applications they show an 8Kw solar array is needed so either that would be much larger than what I need or my calculations are way off.
 
20kw is approximately 415 ah from a 48 volt system a 8k panel array will give you about 160 amps per hour at full output in perfect conditions not a normal daily output figure 70 to 80 percent of that real world I would look at a 12k or larger array minimum
 
That calculator looks pretty worthless.

Assume you mean 20kWh/day.

Math error:

1.66h * 1.2kWh = 1.992kWh

Using typical assumptions, a 20kWh daily use would need 4000W of PV in good solar conditions with full solar exposure from sunrise to sunset with no shading facing due South and tilted horizontally to your latitude. Worse in winter.

Not sure what you mean by 90-160A. Amps are a measure of instantaneous current. 20kWh is a measure of energy. If you meant Ah, system voltage must be known.
 
20kw is approximately 415 ah from a 48 volt system a 8k panel array will give you about 160 amps per hour at full output in perfect conditions not a normal daily output figure 70 to 80 percent of that real world I would look at a 12k or larger array minimum

12kW to get 20kWh/day?

That's gross overkill unless you're at an extreme northern latitude with shitty winter weather.

I'd use the links in line 1 of my signature to get specific insolation data and design for the worst case month.
 
That calculator looks pretty worthless.

Assume you mean 20kWh/day.

Math error:

1.66h * 1.2kWh = 1.992kWh

Using typical assumptions, a 20kWh daily use would need 4000W of PV in good solar conditions with full solar exposure from sunrise to sunset with no shading facing due South and tilted horizontally to your latitude. Worse in winter.

Not sure what you mean by 90-160A. Amps are a measure of instantaneous current. 20kWh is a measure of energy. If you meant Ah, system voltage must be known.

Yes, I'm aware that amps are measure of instantaneous current. Like I said, everything I know about solar is from my experience with RV solar systems. For example, it's common for someone to say that they used 25 amps if their 100 amp hour battery is drawn down by 25 amps.

Am I correct in that amps X volts = watts? Therefore, 20000 watts/120 volts = 166 amp hours would be used, correct? At 220 volts that would be 90 amps hours, right? I know there are losses due to inefficiencies in the inverter but in principal, what am I missing? Would it be 12 volts because that's what the inverter would draw?

In terms of recharging the battery for what was taken out, isn't that Time = Ah / Amps? Wouldn't a 1200 watt array produce 100 amps (1200/12v=100)?

I looked at multiple calculators and they all back that up but I'd like to understand how that is wrong. Here's an article confirming in principal the charge time;

 
Yes, I'm aware that amps are measure of instantaneous current. Like I said, everything I know about solar is from my experience with RV solar systems. For example, it's common for someone to say that they used 25 amps if their 100 amp hour battery is drawn down by 25 amps.

Right, but what if someone has a 12V battery and someone else has a 24V battery. If they both use 25A for 1h, how much energy are they using?

12V * 25A * 1h = 300Wh
24V * 25A * 1h = 600Wh

If you use Ah, your energy consumption in Wh depends on voltage - it must be specified.

Power in = power out

12VDC * 100A = 1200W input
120VAC * 10A = 1200W output

(inefficiencies ignored)

Am I correct in that amps X volts = watts? Therefore, 20000 watts/120 volts = 166 amp hours would be used, correct?

Only if you have a 120V battery.

At 220 volts that would be 90 amps hours, right?

Only if you have a 220V battery.

I know there are losses due to inefficiencies in the inverter but in principal, what am I missing?

A fundamental understanding of Wh.

A Wh is 1W continuously expended over the course of 1h.

20000Wh/24h = 833W - you're burning an average of 833W all the time.

833W / 12V = 69A
833W / 120V = 6.9A

10X the voltage requires 10% the current for the same wattage, i.e., current and subsequently Ah is dependent on voltage.

Would it be 12 volts because that's what the inverter would draw?

I'm not sure how many folks are using 120 and 240V batteries. Expressing AC consumption in amp hours is nothing I've ever seen before. Let's carry that through. A 1200W array can only generate, 1200W/120V = 10A. 1.66h * 10A = 16.6Ah or 10% of what you need. You can't change voltage arbitrarily without changing the results accordingly.

In terms of recharging the battery for what was taken out, isn't that Time = Ah / Amps? Wouldn't a 1200 watt array produce 100 amps (1200/12v=100)?

Yes, but you keep changing voltages. First you use 120V, then you use 12V. When you change voltages, the amps change. That's why Wh is a little easier - it's independent of volts since it's incorporated into the value.

Daily energy: 20kWh
Array: 4kW

20kWh/4kW = 5 hours.

Note how the units carry through. You get to skip Volts and Amps altogether.

I looked at multiple calculators and they all back that up but I'd like to understand how that is wrong. Here's an article confirming in principal the charge time;


You're not understanding the basics, so you're getting the calculator inputs wrong - Wh is consistent across voltages because both V and Ah are in the calculation of Wh.

5A * 12V * 1h = 60Wh
5A * 120V * 1h = 600Wh (10X the energy for the same current due to 10X the voltage)

Back to your consumption:

20000Wh/120VAC = 167Ah
20000Wh/12V = 1667Ah

The two above items are equivalent. If you change voltage, the Amps/Ah has to change accordingly.

You don't get free energy simply by changing your voltage.
 
@sunshine_eggo Thanks for the lengthy and detailed reply. It looks like I had the formulas correct but my fundamental flaw was having the wrong voltage input to understand how many amp hours I would be drawing from the batteries. 1667Ah would be the number that I would be using in that scenario as the batteries are 12v batteries. I originally used 120v or 220v as that would be the voltage that the inverter would be supplying to the house but I had that backwards which is why I wanted to check that and not make an assumption.

I'm going to target having a 6-8kWh array and a larger battery backup then.

 
@sunshine_eggo Thanks for the lengthy and detailed reply. It looks like I had the formulas correct but my fundamental flaw was having the wrong voltage input to understand how many amp hours I would be drawing from the batteries. 1667Ah would be the number that I would be using in that scenario as the batteries are 12v batteries. I originally used 120v or 220v as that would be the voltage that the inverter would be supplying to the house but I had that backwards which is why I wanted to check that and not make an assumption.

I'm going to target having a 6-8kWh array and a larger battery backup then.


I would encourage you to simulate said array at PVWatts (link in my signature) and confirm that winter performance meets expectations.
 
12kW to get 20kWh/day?

That's gross overkill unless you're at an extreme northern latitude with shitty winter weather.

I'd use the links in line 1 of my signature to get specific insolation data and design for the worst case month.
Yes living up north with snow and clouds short pv days and not having any grid to fall back on at our mountain home I may over panel more than most but it’s not sunny az or ca and I don’t want to be running a gen set all winter and panels are pretty cheap compared to a larger battery bank
 
Yes living up north with snow and clouds short pv days and not having any grid to fall back on at our mountain home I may over panel more than most but it’s not sunny az or ca and I don’t want to be running a gen set all winter and panels are pretty cheap compared to a larger battery bank

Sure, they're "cheap," but where do you draw the line?

12kW may be the correct answer for you, but OP is inquiring on an existing dwelling that almost certainly has grid access.

Worst case in Stafford Springs, CT is 2.41h for December, and that's primarily due to weather. Even with 12kW, hourly data shows a 12kW array would fail to generate 20kWh 14 out of 31 days with 5 of those 14 days consecutive, so even a 12kW array wouldn't do it. A 20kW array would still miss 9 of 31 days. A 30kW would still miss 4 days. The bottom line is that certain locations demand a backup to PV.

We're also assuming the 20kWh average applies to every month. Winter may be a low if heat is not electric. May be even higher if it is.

Best to analyze one's specific situation as I recommended above.
 
I would encourage you to simulate said array at PVWatts (link in my signature) and confirm that winter performance meets expectations.
That was my initial expectation - to have a system that meets the minimum available daylight hours. In my case that's 2.5-3 hours. Based on that link, a 6kWh system would be enough on an annualized basis but a 9-10kWh system would be needed to provide 100% of the usage.
 
That was my initial expectation - to have a system that meets the minimum available daylight hours. In my case that's 2.5-3 hours. Based on that link, a 6kWh system would be enough on an annualized basis but a 9-10kWh system would be needed to provide 100% of the usage.

Yeah. 8kW seems to be the minimum to meet 20kWh average in December, but make sure you look at the hourly data. Download the CSV and look at the days that don't generate 20kWh. Based on the referenced location, 8kW would fail to generate 20kWh 19 out of 31 days.

Best if you can get monthly consumption and do the analysis. If all your heat is from fire, you may need less than 20kWh/day in December.
 
Yeah. 8kW seems to be the minimum to meet 20kWh average in December, but make sure you look at the hourly data. Download the CSV and look at the days that don't generate 20kWh. Based on the referenced location, 8kW would fail to generate 20kWh 19 out of 31 days.

Best if you can get monthly consumption and do the analysis. If all your heat is from fire, you may need less than 20kWh/day in December.
Yeah, I'm not too worried about winter as I would be using wood. I'm also going based on the previous owners energy usage. I plan on swapping out some higher energy consuming appliances to reduce the load. I figure I'll get more bang for the buck that way.
 
Yeah, I'm not too worried about winter as I would be using wood. I'm also going based on the previous owners energy usage. I plan on swapping out some higher energy consuming appliances to reduce the load. I figure I'll get more bang for the buck that way.
It’s a lot cheaper to conserve than to generate power so if you can cut that 20k daily down it will save you some money but look at costs of some of these appliances vrs panels some are not the best bang for the buck a lot of research is needed to really understand the cost
 
Back
Top